The I Ching became more important still with the additions made by or because of Confucius. His influence is seen in both The Commentary on the Decision (which he may have written) and The Commentary on the Images (Wilhelm 1950, xxxix-xxx).
The literary machine in Western culture-as a physical machine, as well as a procedural way of generating texts-seems to have been devised in 1274 C.E. by the Catholic alchemist, mystic, and philosopher Ramon Llull (1232-1316). Although his system is too elaborate to explain fully here, he related nine letters of the alphabet each to a quality, a relation, a question, a subject, a virtue, and a vice, and then created a system to, as Llull scholar Andrew Bonner (1997) writes, "combine these theological, scientific and moral components to produce arguments." One page of his Ars Generalis Ultima and the shortened version of that work, the Ars Breva, as they were circulated in Hull's time, typically included a physical text-generating machine that allowed three-letter combinations to be devised. There were three circles, each with the letters of the alphabet drawn along the circumference at even intervals. Bonner describes the way these parts were made into a machine: "In medieval manuscripts, the outside circle is normally drawn on the page, and the two inner ones are separate pieces of parchment or paper held in place on top of it by a little piece of string, permitting them to rotate in relation to each other and to the larger circle."
Although the I Ching is more widely known in the West today than is Hull's machine, the latter was also quite influential. Llull, known as "Doctor Illuminatus," had a tremendous following. He wrote approximately three hundred books, with the ones detailing his text-generating machine becoming some of the most important and influential. Chairs were established in universities in Barcelona and Valencia after his death to further his studies. When Gottfried Wilhelm Leibniz attempted to create the first mechanical calculator, his work was informed by Llull's writing and the description of his machine (Bonner 1997). Despite being beatified (usually the first step toward sainthood), Llull was not canonized. His ideas-which, as an article in The Catholic Encyclopedia explains, threatened "breaking down the distinction between natural and supernatural truth"-were condemned by the Church (Turner 1911).
The next important literary machine that was formulated in Western history was described only in satire, by a riddler who has been mentioned already. It appeared in Gulliver's Travels. A computer historian writes: "In 1726 Johnathan Swift published a description of a wonderful machine, made of equal parts of irony, sarcasm, and mockery, that would automatically write books on all the arts and sciences" (Weiss 1985, 164). Swift described Gulliver visiting Balnibarbi on his third voyage and surveying a range of absurd academic endeavors in the Grand Academy of Lagado. On meeting the first professor Gulliver sees
a Frame ... twenty Foot square, placed in the middle of the Room. The Superficies was composed of several Bits of Wood, about the bigness of a Die, but some larger than others. They were all linked together by slender Wires. These bits ofWood were covered on every Square with Paper pasted on them, and on these Papers were written all the Words of their Language, in their several Moods, Tenses, and Declensions, but without any Order. (Swift [1726] 1735, pt. 3, chap. 4)
By turning iron handles at the edge, the professor's forty students bring different words into view. Then, as most of them look over the results and report any series of words that might make a sentence, a few serving as scribes write down these generated texts. Although offered in jest, Swift's hypothetical literary machine has been remembered and invoked in the information age. Weiss (1985) explains that the machine's "purpose, the claims of its illustrious professor-inventor, his call for public funding, and the operation of the device by students clearly classify it as an early attempt at artificial intelligence and have caused it to be cited often as typical of this discipline" (164).
The generation of secular texts by procedure was done early in the last century to interesting effect, as William Burroughs tells it. He wrote that at a "rally in the 1920's Tristan Tzara the man from nowhere proposed to create a poem on the spot by pulling words out of a hat. A riot ensued wrecked the theatre" (Burroughs 2003, 90). Romanian-born Tzara, who was a founder of Dada and who went on to become an important part of another French movement, surrealism, around 1930, is credited with being the originator of the "cut-up" technique of random recombination of text. Tzara's perspective on the production of language could not have been more different than that of Llull, who sought to systematically establish important religious and philosophical truths. "Everything one looks at is false," Tzara wrote in the "Dada Manifesto," the first of seven that he wrote. "I do not consider the relative result more important than the choice between cake and cherries after dinner" (Tzara 1951).
Brion Gysin (1982) followed in that anti-tradition three decades later, without at first realizing it:
While cutting a mount for a drawing in room 25, I sliced through a pile of newspapers with my Stanley blade and thought of what I had said to Burroughs some six months earlier about the necessity for turning painters' techniques directly onto writing. I picked up the raw words and began to piece together texts. (51)
The technique, developed in the summer of 1959, was used to write Minutes to Go (Burroughs 2003, 90). (Another interesting connection between computing and literature is that Burroughs was the grandson of the inventor of the Burroughs Adding Machine, who was also named William S. Burroughs; the novelist was heir to the fortune created by this early computer.) Burroughs felt this cut-up technique was a step toward the more complete automation of language. Burroughs wrote that the "cut-ups, permutations and tape recorder experiments carried out by Brion Gysin are aimed ... toward making words talk on their own" (qtd. in Gysin 1982, xi).
France was the site of another important event in the history of the literary machine. On 24 November 1960 the Oulipo was founded by Francois Le Lionnais and Raymond Queneau. The group was dedicated not to the creation of literature, but to the creation of methods or ways of creating literature-hence the word "potential" in the name. The literary machine, a form of potential literature, was certainly a proper object of Oulipian study. The group's formulations included algorithms to be applied to an existing text (e.g.,Jean Lescure's N + 7 rule for replacing each noun with the seventh noun after it in a dictionary (Mathews and Brotchie 1998, 198)) and constraints on the authorship of a text (e.g., the text must be a palindrome; letters and numbers must read the same forward and backward). The most succinct definition of the group was provided at a 5 April 1961 meeting by Queneau: "rats who construct the labyrinth from which they propose to escape" (Lescure 1986, 37; Mathews and Brotchie 1998, 201).
One physical device for generating texts was Queneau's 1961 Cent mille milliard de poemes (100, 000, 000, 000, 000 Poems), which took the form of a book with ten sonnets (each of the usual fourteen lines) bound one in front of the other and with each line cut so that it could be "turned," like a page, separately. Any one of ten lines could thus be selected for each position. An entry in Oulipo Compendium explains: "The rhyme scheme of the sonnets is uniform; grammatical correctness is assured no matter what sequence of lines occurs" (Mathews and Brotchie 1998, 14). Hence there are 10" possible poems in the book, which would take (by Queneau's calculation) more than 190 million years to read. This potential sonnet has, amazingly, been translated into English twice-by Stanley Chapman and by John Crombieand it has been implemented both on computers and as a more stereotypical machine, with gears and rollers (Mathews and Brotchie 1998, 177-178). Queneau also devised the first Choose Your Own Adventure type of story, his 1967 "Un conte a votre facon" (which has been translated as "A story as you like it" and "Yours for the telling").This short work has twenty-one possible text segments; the reader is asked to make a choice after the first one is read, and after each subsequent one, up to one of the possible endings. Warren Motte explains that it was "inspired by the presentation of the instructions given to computers, and by programmed teaching" (1997, 156). The juvenile fiction series
named Choose Your Own Adventure, of similar inspiration, began in 1979 with Edward Packard's The Cave fTime-too late to influence the early stages of interactive fiction, and in fact likely to have been at least vaguely inspired by actual computer programs, including very early interactive fiction. It was one of several series of childrens' books (in many languages) that asked readers to choose the next step to take after each page or so of text. There were more than two hundred such books published in the two main Bantam series, Choose Your Own Adventure and Choose Your Own Adventure forYounger Readers. The Cave ofTime was itself made into a graphical adventure game by Bantam Software and published in 1985.
One book-length antecedent to the Choose Your Own Adventure was Julio Cortazar's 1963 Rayuela (translated as Hopscotch, 1966), a novel that has "expendable chapters" and offers instructions for how to read along two different paths. In one reading sequence, a scene with a very different emotional tone (such as one depicting a rape, narrated by a rapist) is placed between two others; in the others this scene is omitted and the reading is quite different.
Another earlier antecedent was the even more reconfigurable Composition no. 1 by Marc Saporta, published in France in 1961 and translated into English by Richard Howard in 1963. This book-not a codex but 150 loose, unnumbered pages in a box-invites the reader to shuffle the pages and read them in any order. According to the text printed on the box, the 149 texts on the loose pages (one is a title page) tell the story of X, who has an affair, has a disastrous marriage, gambles, serves in the French army during the occupation of Germany, rapes a girl, and is in an auto accidentthough not necessarily in that order. Each text describes something coherent about a particular moment in X's life, providing enough context (often repeated in other, similar texts) to make the moment intelligible. Some of these events are fixed as having happened when X was a child or an adult, but there is ambiguity as to the exact sequence of some events; causation, in several cases, is also unclear, as is the exact nature of some events. A theft that X committed may have involved his stealing an envelope of money from his workplace, or it may have been an envelope of names that, as a member of the Resistance, he stole from the Germans.
The text on the box suggests that a particular reading of the pages, in a particular order, will resolve the events of the story into an order, and it suggests that this order of events will change depending upon how the pages are sorted. It does not seem as though a particular shuffling of the deck will actually render any the ambiguities of the potential story certain, however. A particular reading will allow the reader to ponder the elements of the story in a different order and to try to resolve them along the way. During the process, a different set of post hoc, propter hoc fallacies will inevitably be imagined by the reader, and these will suggest different connections among events, but by the time the reading is finished it will be hard to still believe in any of them. As J. David Bolter (2001) writes, while "Saporta's experiment ... seems to position his work as an inevitable, final step in the exhaustion of printed literature," it seems, even if it is indeed offered as that, to also serve as "a bridge to the electronic medium" where the author constructs narrative possibility rather than narrative itself (150). It is in fact left to the reader of Composition no. I to assemble the pages into a particular telling, and to realize a narrative from the deck of texts provided; although the ordering of events may never be certain, this hardly prevents the reader from associating meaning with and investing emotion in the events that did take place and the characters who took part in them.
Certain literary machines were implemented on computers in the United States before Adventure. Dale Peterson (1983) writes that "Louis Milic, an English professor at Cleveland State University, may have created the first computer poetry. In 1963, Milic programmed a computer to generate absurd English sentences" (138). Milic then refined his program to allow different poems to be generated with the same syntax. This effort had been anticipated by an attempt in England to generate (amusing and nonsensical) physics essays in the 1950s (141). Brion Gysin (1982) himself experimented with computer-generated texts with the help of a mathematician. The computer's use in the generation of poems and stories has its own interesting history. These are only few of the experiments that preceded the first work of interactive fiction; some of what followed included attempts at machine generation of stories (Meehan 1980), a thread of computational and literary endeavor that differs from the creation of interactive systems.
An important early literary machine of another sort was Theodor Nelson's 1970 Labyrinth, which Nelson has said was the first publicly accessible hypertext. Labyrinth was a hypertext catalog of the 1970 Software exhibition at the Jewish Museum. The exhibition, which proved controversial, included conceptual art and unusual applications of technology. The catalog was installed there on a PDP-8 with some difficulty, and it ran for the last month of the exhibition (Shanken 1998). Although this work was nonfiction, and although it was never as widely known as Nelson's influential books were, Labyrinth was pioneering in offering a reconfigurable text on a topic outside the usual sphere of mainframe computing.
Looking back from twentieth-century text-generating experiments to Llull and the I Ching, it is important to recognize that, as Peterson (1983) reminds us,
When computer programmers and a few poets first produced machine poems based on much the same principle as that of Tristan Tzara ... they did not see themselves as the odious speculators in Swift's Academy of Lagado, nor as poets in the Dada tradition. Mostly, they were light-hearted experimenters, trying to discover the word-manipulation possibilities of a new machine. (137)
Nevertheless, those working with computers were, and continue to be, influenced by early work on literary machines. Versions of early literary machines that have been implemented on digital computers, citations of these machines in computing literature, and presentations about these early innovations at computer-oriented conferences demonstrate such influence rather directly. In any case, this brief history should indicate something about the possibilities of computer literature, since it shows the wide range of purposes to which literary machines had already been put, before interactive fiction was devised: prophetic, theological, satirical, nihilistic, and playful.
DUNGEONS AND DRAGON S
The first successful fantasy role-playing game was formulated in the early 1970s in Lake Geneva, Wisconsin, and first published in 1974. Gary Gygax published a precursor to Dungeons and Dragons in a magazine he ran for the Castle and Crusade Society, part of the International Federation of Wargaming. Gaming of this sort involved 1:20 scale figurines and maps. With Dave Arenson, Gygax then developed a new sort of campaign incorporating character classes ("Heroes or Wizards"), experience points, and a dungeon maze. Spells and monsters were incorporated into this first Dungeons and Dragons campaign, Greyhawk (Gygax 1980). In summary,
D&D is an open-ended game in which the players assume the roles of characters in a story and can have those characters attempt any action whatsoever. The game is controlled by a gamesmaster, who uses tables, dice, and personal judgment to decide on the effect of a character's efforts. (McGath 1984, 6)
This "gamesmaster" is called the dungeon master. The players say what their characters, the "player characters," do within the "world" of the campaign. Dice are rolled to assign ability points to player characters initially, and dice are employed to help determine the outcome of combat and other encounters. Over the course of many adventures, the members of a party advance in level and become more powerful-and the dungeon master devises new challenges for them.
Figurines and dice are not as central to Dungeons and Dragons as is sometimes thought. Miniatures, although important to certain war games that preceded Dungeons and Dragons, are seldom used in playing Dungeons and Dragons. For the most part they are simply merchandise, and are even advertised as "collectibles" rather than as essential components of the game. Dice, on the other hand, are frequently used in a typical game. The twenty-sided die and othe
r dice of unusual shape have become iconic. Dice are used to introduce unpredictability; based on the roll of a die, events may transpire that were not anticipated, foiling even the dungeon master's plans. It is not unheard of for people to play without dice, however. Chance occurrences are simply one element within a framework of problem solving, role playing, and exploration of an imagined, fantastic world. Consider this player's perspective, printed in Dragon magazine:
The dreamer's art, the ability to cut loose from the restraints of reality and touch new shores and lives, is the essence and lure of D + D. It is the challenge of pitting one's skills and common sense against a strange and sometimes hostile universe where death awaits with open arms. (Filmore 1980)
The extent to which Dungeons and Dragons is inspired by J. R. R. Tolkien's work has frequently been misunderstood and overstated. Although Tolkien's books are obviously important to the fantasy genre and were influential on many particular Dungeons and Dragons campaigns undertaken by groups of players, Tolkien can sometimes seem the single straw that those unfamiliar with fantasy and adventure writing grasp at when trying to understand where this game came from and how to situate it vis-a-vis literature. Recent film releases have not helped correct this misperception. Many writers have assumed that Tolkien's books are the basis for the gamefor instance, Christian opponents of the game have assumed this (Weldon and Bjornstad 1984, 49). This confusion is not restricted to outside observers. Some players of Dungeons and Dragons have similarly thought that the game was Tolkien-inspired and Tolkien-centered. The official statement from TSR, the makers of the game, in response to letters from such players is that "D&D was not written to recreate or in any collective way simulate Professor Tolkien's world or beings ... this system works with the worlds of R. E. Howard, Fritz Leiber and L. S. de Camp and Fletcher Pratt much better than that of Tolkien" (Kuntz 1980). Middle Earth was created for the sake of a single adventure, as TSR saw it, while other fantasy literature, more closely related to Dungeons and Dragons, has been set in a world rich with continual adventure.
Twisty Little Passages Page 9